PhP

O

MD(

YF

a

b

c

d

SM

Peetiopscioli©

K

1

e

h1

rotist, Vol. 166, 363–373, July 2015ttp://www.elsevier.de/protisublished online date 28 May 2015

RIGINAL PAPER

olecular Phylogeny of the Widelyistributed Marine Protists, Phaeodaria

Rhizaria, Cercozoa)

asuhide Nakamuraa,1, Ichiro Imaia, Atsushi Yamaguchia, Akihiro Tujib,abrice Notc, and Noritoshi Suzukid

Plankton Laboratory, Graduate School of Fisheries Sciences, Hokkaido University,Hakodate, Hokkaido 041–8611, JapanDepartment of Botany, National Museum of Nature and Science, Tsukuba 305–0005,JapanCNRS, UMR 7144 & Université Pierre et Marie Curie, Station Biologique de Roscoff,Equipe EPPO - Evolution du Plancton et PaléoOcéans, Place Georges Teissier, 29682Roscoff, FranceInstitute of Geology and Paleontology, Graduate School of Science, Tohoku University,Sendai 980–8578, Japan

ubmitted January 1, 2015; Accepted May 19, 2015onitoring Editor: David Moreira

haeodarians are a group of widely distributed marine cercozoans. These plankton organisms canxhibit a large biomass in the environment and are supposed to play an important role in marinecosystems and in material cycles in the ocean. Accurate knowledge of phaeodarian classification ishus necessary to better understand marine biology, however, phylogenetic information on Phaeodarias limited. The present study analyzed 18S rDNA sequences encompassing all existing phaeodarianrders, to clarify their phylogenetic relationships and improve their taxonomic classification. The mono-hyly of Phaeodaria was confirmed and strongly supported by phylogenetic analysis with a larger dataet than in previous studies. The phaeodarian clade contained 11 subclades which generally did notorrespond to the families and orders of the current classification system. Two families (Challengeri-dae and Aulosphaeridae) and two orders (Phaeogromida and Phaeocalpida) are possibly polyphyleticr paraphyletic, and consequently the classification needs to be revised at both the family and order

evels by integrative taxonomy approaches. Two morphological criteria, 1) the scleracoma type and 2)ts surface structure, could be useful markers at the family level.

2015 Elsevier GmbH. All rights reserved.

ey words: Phaeodaria; Cercozoa; Rhizaria; 18S rDNA; single-cell PCR.

Corresponding author; fax +81 138 40 5542-mail y.nakamura@fish.hokudai.ac.jp (Y. Nakamura).

ttp://dx.doi.org/10.1016/j.protis.2015.05.004434-4610/© 2015 Elsevier GmbH. All rights reserved.

364 Y. Nakamura et al.

Introduction

Phaeodarians are marine unicellular protists.This group is defined by 1) the presence of a“central capsule” containing one or several nuclei,2) a mass of brown aggregated particles named“phaeodium”, and 3) a hollow siliceous skeletoncalled “scleracoma”, in most species (Howe et al.2011; Nakamura and Suzuki 2015; Takahashi andAnderson 2000). The size of the scleracoma varieswith the different families and ranges from severalhundred micrometers (e.g. Challengeriidae) to afew millimeters (e.g. Tuscaroridae). The type of thescleracoma and its surface structure are importanttaxonomic characters to classify these organismsat the order and family levels. Phaeodarians havelong been considered as a member of Radiolariabut based on molecular analyses (Nikolaev et al.2004; Polet et al. 2004; Sierra et al. 2013), theyare now classified as a subclass in the classThecofilosea of the phylum Cercozoa (Adl et al.2012; Howe et al. 2011).

Phaeodarians are heterotrophic plankton orga-nisms, which are distributed in the world ocean fromthe surface to the deep sea (Nakamura and Suzuki2015). Despite their wide range of distribution, thisplankton group has attracted little attention frommarine biologists (Nakamura et al. 2013). Theirabundance has been very likely underestimated,because their scleracoma is fragile and breakseasily when samples are collected by planktonnets. Yet, careful examination of the zooplank-ton community revealed that these protists can benumerous and their high abundance has occasion-ally been reported. For instance, the vertical flux ofphaeodarians was higher than that of polycystineradiolarians in the Panama Basin (Takahashi andHonjo 1983) and in the California coast (Gowingand Coale 1989). The combined biomass of theAulosphaeridae, Sagosphaeridae, Aulacanthidaeand Coelodendridae was estimated to contribute2.7–13.7% of the total metazoan biomass in the150–1000 m layer of the western North Pacific(Steinberg et al. 2008). The proportion of anAulacanthidae species, Aulographis japonica, withrespect to the total zooplankton biomass is 22.3%in 250–3000 m layer in the Sea of Japan, andthis percentage is the second largest, followingthat of copepods (Nakamura et al. 2013). Con-sidering that they possess siliceous skeletons andoccur occasionally at high biomass, this planktongroup can play an important role in ecosystemslocally and have a significant impact on the sil-ica cycle of the ocean. Their importance in thecarbon cycle is also reported from the Northeast

Atlantic (Lampitt et al. 2009). Cercozoans, includ-ing phaeodarians, have recently been consideredimportant players in the material cycles and foodwebs of the ocean (Howe et al. 2011). Researchon environmental DNA revealed that Cercozoarepresented a high percentage of the 18S rDNAsequences obtained from sea-floor sediments ofthe Arctic and the Southern Ocean (Pawlowskiet al. 2011). Consequently, improving our knowl-edge on phaeodarians will be important for theentire field of marine biology. Accurate knowl-edge of phaeodarian classification is critical tobetter assess their genetic diversity and ecology. Inthe comprehensive systematics framework estab-lished by Haeckel (1887), phaeodarian specieswere grouped into families and orders according totheir morphological similarities. Since then, someauthors amended this classification, but the rela-tionship between the families remains uncertain(Cachon and Cachon 1985; Campbell 1954; Klingand Boltovskoy 1999; Takahashi and Anderson2000) and has never been examined from themolecular point of view. Comprehensive molecu-lar studies on phaeodarians are difficult to conduct.They cannot be cultured, and therefore their DNAhas to be extracted from individual specimens col-lected and sorted from the environment. Somephaeodarian species, however, live in the deep sea,and it is hard to obtain specimens. In addition,efficient phaeodarian-specific primers do not exist,increasing the risks of amplifying and sequencingcontaminating organisms attached to or ingestedby the Phaeodaria.

Yet, an integrative taxonomy approach merg-ing both morphological and molecular informationhas recently been applied successively to otherunicellular and shell-bearing rhizarian taxa suchas Polycystines, Acantharia and Foraminifera (e.g.Decelle et al. 2012; Kunitomo et al. 2006;Pawlowski et al. 2013). Such an approach,highlighting match and mismatch between mor-phology and phylogenetic analysis, contributedto an overall better understanding of the classi-fication and evolutionary patterns among thesegroups.

The information on phaeodarian 18S ribosomalDNA currently exists for only 7 out of ca. 200species, 18 families and 7 orders recognized inthe latest phaeodarian classification (Nakamuraand Suzuki 2015; Takahashi and Anderson 2000),and discrepancies between morphological classi-fication and analysis of these partial 18S rDNAwere pointed out (Nakamura et al. 2013). Here,we designed new phaeodarian-specific primersand analyzed the whole 18S rDNA sequence

Molecular Phylogeny of Phaeodaria 365

of extant single-cell specimens of phaeodarianspecies, covering all orders recognized in the cur-rent classification.

Results

A total of 39 phaeodarian specimens belongingto 26 distinct taxa (essentially distinct species)were collected from 0 m to 3 000 m depth, and

their sequences were obtained by single-cell PCR.Detailed information and micrographs of analyzedspecimens are shown in Table 1 and Figure 1,respectively.

The sequences from all phaeodarian specimensformed a single clade far separated from otherThecofilosea taxa (Cryomonadida, Ebriacea andPseudodifflugia), and the monophyly of this cladewas strongly supported by bootstrap values andBPP (Fig. 2). The phaeodarian clade consisted of

Table 1. List of phaeodarian sequences used for the phylogenetic analysis in this study. the specimens arecoded according to the sampling station: the wesern North Pacific (WNP); the eastern north Pacific (ENP); theEast China Sea (ECS); the Sea of Japan (SJ); the Mediterranean Sea (MS).

Taxon Accession no. Specimen Sampling Depth Samplingname station range (m) date

Protocystis vicina AB998884 WNP1 WN1 250–500 Aug. 2013Protocystis vicina AB998885 WNP2 WN1 250–500 Aug. 2013Challengeriidae sp. 1 AB998886 WNP3 WN2 100–150 Aug. 2013Porospathis holostoma AB998887 WNP4 WN2 150–250 Aug. 2013Challengeriidae sp. 2 AB998888 WNP5 WN2 500–750 Aug. 2013Challengeron bethelli AB998889 WNP6 WN2 500–750 Aug. 2013Protocystis thomsoni AB998890 WNP7 WN2 500–750 Aug. 2013Challengeron tizardi AB998891 WNP8 WN2 750–1000 Aug. 2013Challengeron tizardi AB998892 WNP9 WN2 1500–2000 Aug. 2013Challengeron tizardi AB998893 WNP10 WN2 1500–2000 Aug. 2013Entocannula infundibulum AB998894 WNP11 WN2 1500–2000 Aug. 2013Porospathis holostoma AB998895 WNP12 WN2 1500–2000 Aug. 2013Protocystis harstoni AB998896 WNP13 WN2 1500–2000 Aug. 2013Protocystis murrayi AB998897 WNP14 WN2 1500–2000 Aug. 2013Tuscaretta belknapi AB998898 ENP1 EN1 0–300 Aug. 2012Coelodendrum furcatissimum AB998899 ENP2 EN2 0–1000 Aug. 2012Sagoscena sp. 1 AB998900 ENP3 EN2 0–1000 Aug. 2012Tuscaretta belknapi AB998901 ENP4 EN3 0–1000 July 2012Tuscaretta sp. 1 AB998902 ENP5 EN4 0–1000 July 2012Auloceros arborescens AB998903 ENP6 EN5 0–1000 July 2012Coelodendrum furcatissimum AB998904 ENP7 EN5 0–1000 July 2012Conchariidae sp. 1 AB998905 ECS1 E1 0–150 May 2013Challengeron radians AB998906 ECS2 E2 200–500 May 2013Conchariidae sp. 2 AB998907 ECS3 E2 200–500 May 2013Challengeron diodon AB998908 ECS4 E3 650–800 May 2012Challengeron radians AB998909 ECS5 E3 650–800 May 2012Challengeron willemoesii AB998910 ECS6 E3 650–800 May 2012Challengeron willemoesii AB998911 ECS7 E3 650–800 May 2012Protocystis xiphodon AB998912 ECS8 E4 0–150 May 2012Protocystis xiphodon AB998913 ECS9 E4 0–150 May 2012Aulographis japonica AB998914 SJ1 S1 250–750 Apr. 2013Auloscena sp. 1 AB998915 SJ2 S1 250–750 Apr. 2013Protocystis vicina AB998916 SJ3 S1 250–750 Apr. 2013Aulographis japonica AB998917 SJ4 S2 250–750 Apr. 2012Auloscena sp. 1 AB998918 SJ5 S3 500–750 Mar. 2013Aulacantha scolymantha AB998919 MS1 M1 0–40 Nov. 2012Medusetta parthenopaea AB998920 MS2 M1 0–40 Nov. 2012Phaeodina sp. 1 AB998921 MS3 M1 0–40 Nov. 2012Phaeodina sp. 1 AB998922 MS4 M1 0–40 Nov. 2012

366 Y. Nakamura et al.

Figure 1. Light micrographs of phaeodarian specimens analyzed in this study. The information on each spec-imen is presented in Table 1. cc: central capsule; ph: phaeodium; sc: scleracoma.

Molecular Phylogeny of Phaeodaria 367

Figure 1. (Continued)

11 subclades (A–K) defined based on support val-ues and morphological features (Fig. 3). SubcladeA contains the highest number of individuals (n =18). Subclades A, E, F, G, H and K were composed

of sequences from distinct sampling locations. Forinstance, subclade E contains the specimens fromthe Mediterranean Sea (MS), the Sea of Japan (SJ)and the eastern North Pacific (ENP).

368 Y. Nakamura et al.

Figure 2. Phylogenetic tree of phaeodarians andother cercozoans based on 18S rDNA alignmentsand maximum-likelihood (ML) method. Note that thephaeodarian clade is presented with 11 subclades(A–K) detected according to the support values andmorphological features. The details of this cladeare shown in Figure 3. Numbers at nodes indicateBayesian posterior probabilities (BPP) and bootstrapsupport values of neighbor-joining (NJ) and ML meth-ods (BPP/NJ/ML). Only values higher than 0.5 pp and50% are shown.

The 11 subclades and the families classified byTakahashi and Anderson (2000) generally did notmatch well, and disagreement was seen for thefamilies Aulosphaeridae, Sagosphaeridae, Chal-lengeriidae and Aulacanthidae (Fig. 3). Sagoscenasp. 1 (ENP3) of Sagosphaeridae formed the sub-clade E along with two Aulosphaeridae species,being nested one into each other. Challengeri-idae and Aulacanthidae were scattered acrossthe phylogenetic tree. Most of the challengeriidphaeodarians appeared in subclade A, but twospecimens of Challengeron willemoesii Haeckel(ECS6 and ECS7) formed the subclade B clearlydistinct from subclade A with high support values(1.0/100/100, Fig. 3). The sequences of C. diodon(Haeckel) (ECS4 and NA (AB218765)) belongedto subclade H, distant from subclade A, togetherwith the Medusetta parthenopaea Borgert (MS2) ofMedusettidae (Fig. 3). Two Aulacanthidae species,Aulographis japonica Nakamura, Tuji and Suzuki(SJ1, SJ4 and SJ (AB820365)) and Aulocerosarborescens Haeckel (ENP6) formed a single sub-clade K, whereas Aulacantha scolymantha Haeckel(MS1 and MS (AY266294)), which also belongs toAulacanthidae, composed subclade I whose exactposition compared to other subclades is unclear.

The orders of the current classification did notcorrespond to the phylogenetic analysis overall

(Fig. 3). The order Phaeogromida was scatteredacross subclades A, B and F, and the order Phaeo-cystida has representatives in subclades I andK. The order Phaeosphaerida, composed of twofamilies Sagosphaeridae and Aulosphaeridae (sub-clade E), turned out to be nested within the orderPhaeocalpida (subclades C and D).

Discussion

Comparison Between the PhylogeneticTree and the Current Classification

Previous studies analyzed partial 18S rDNA ofphaeodarian species (n=6) belonging to 5 dis-tinct families and orders, and reported that theirsequences formed a monophyletic clade outside ofRadiolaria but within cercozoans (Polet et al. 2004;Yuasa et al. 2006). In this study, the monophyly ofPhaeodaria within Cercozoa is further confirmedbased on a larger data set, including 29 distinctspecies distributed over all 7 orders. Phaeodariansare now classified as a member of the class The-cofilosea, together with Cryomonadida, Ebriaceaand Pseudodifflugia (Adl et al. 2012; Howe et al.2011). However, these Thecofilosea taxa did notform a monophyletic group in the 18S rDNA tree(Fig. 2), and their phylogenetic relationship couldnot be confirmed in the present study. For bet-ter clarifying the phylogenetic relationship betweenphaeodarians and other cercozoans, it would benecessary to add more taxa whose 18S rDNAsequences are not yet available (e.g. Ventricleftida).

The 11 subclades in the 18S rDNA phyloge-netic reconstruction did not well correspond to thefamilies and the orders in the current classifica-tion system (Fig. 3). The family Challengeriidaeis possibly a polyphyletic group, and the familyAulosphaeridae would be a paraphyletic group.As for ordinal level categorization, Phaeogromidaturned out to be polyphyletic. The branching pat-tern of subclades C, D and E suggests that theorder Phaeocalpida is a paraphyletic group. Con-sequently, the current classification system is notsupported by the 18S rDNA phylogeny and needsto be reconsidered.

Comparison Between the PhylogeneticTree and the Morphological Characters

Previous studies classified phaeodarian familiesaccording to morphological characters, especiallybased on 1) the scleracoma type and 2) the sur-face structure of the test (Cachon and Cachon

Molecular Phylogeny of Phaeodaria 369

Figure 3. The phaeodarian clade in the phylogenetic tree inferred by maximum-likelihood (ML) method, basedon 46 sequences of 18S rDNA. The outgroups are shown in Figure 2. Note that the 11 subclades (A–K) aredetected based on the support values and morphological features. Each specimen is coded according to itssampling station as shown in Table 1, and species retrieved from NCBI database are presented with theiraccession numbers. Numbers at nodes indicate Bayesian posterior probabilities (BPP) and bootstrap supportvalues of neighbor-joining (NJ) and ML method (BPP/NJ/ML). Only values higher than 0.85 pp and 50% areshown. The families and orders are based on Takahashi and Anderson (2000).

1985; Campbell 1954; Haeckel 1887; Kling andBoltovskoy 1999; Takahashi and Anderson 2000).The 11 subclades were generally congruent withthese two morphological criteria in this study(Fig. 3). The scleracoma of subclades A, B, C, D,F and H formed relatively firm “test”, but display-ing distinction between each other: cape-shaped orovate test possessing smooth or small-pored sur-face (A and B), spherical test with “tubular spines”and the surface of “trizonal meshwork” (C), testof smooth surface with short “oral spines” andlong “aboral spines” (D), campanulate test pos-sessing developed “peristome” with “arches” (F),and test composed of two “valves” with large-pored surface (H). The scleracoma of subclade Ecorresponds to a sphere of geometric meshwork

called “lattice-shell”. The phaeodarians of subcladeG possess scleracoma composed of “inner shell”and numerous branch-like structure called “styles”.The specimens included in subclades I and Khave scleracoma called “spherical veil”. Specimensbelonging to subclade J do not possess scleracoma(Figs 1, 3).

Thus, the scleracoma type and the surface struc-ture of the test would be valuable characters forthe family-level classification. Future studies shouldrearrange phaeodarian families and orders accord-ing to these two criteria and the branching patternof the phylogenetic tree. The combination of bothdetailed morphological examination and molecularphylogeny helped identifying new characters for thetaxonomic classification of Phaeodaria. Such an

370 Y. Nakamura et al.

integrative approach has already been proven valu-able for a number of taxa, including some amongRhizaria (e.g. Acantharia, Decelle et al. 2012).

Taxonomical Consideration on theFamilies Challengeriidae andMedusettidae

The subclade A contains 18 specimens belongingto 3 different challengeriid genera: Challengeron,Protocystis and Entocannula. These 3 genera wereerected according to the difference in their mor-phological characteristics. Phaeodarians of thesegenera, however, did not make subclades clearlydistinct from each other, and there is a possibilitythat the current classification within the family Chal-lengeriidae is artificial. In order to further clarify theirintra-family relationship, it is necessary to analyzehigher resolution genetic maker regions, such asthe D1 and D2 regions of the 28S rDNA or ITS.

The genus Challengeron contains ca. 13species among which 5, including the type speciesC. bethelli (Murray) (WNP6), were analyzed inthis study (Campbell 1954). The scleracoma ofC. willemoesii (ECS6 and ECS7) forms a firm“test”, similar to other challengeriid phaeodarianscomposing subclade A (Fig. 1). This species has atest with the surface made of a regularly arrangedamphora-shaped structure, which is common toother challengeriid species, except C. diodon(ECS4) (Takahashi 1991). Consequently, thesurface structure does not distinguish this speciesfrom other challengeriids examined in this study.Challengeron willemoesii has a developed “peris-tome” with four “oral spines” (Fig. 1). This complexperistome structure is not seen in other challen-geriids belonging to subclade A, and this charactermay be a morphological criterion distinguishingthis species from other challengeriids.

The fact that C. diodon (ECS4) appeared insubclade F together with Medusetta parthenopaea(MS2) of the family Medusettidae suggests that theformer is phylogenetically closer to the medusettidsthan other challengeriids. Morphological differ-ences of C. diodon from other challenderiids havepreviously been noticed: 1) a dimpled surface; 2)clusters of small pores associated with individualamphorae (inner shell surface); and 3) thin anddelicate amphorae (Takahashi 1991). This authoralso mentioned that the overall appearance of thisspecies was closer to that of medusettid. The twospecies composing subclade F have one “aboralspine” and the developed “peristome” with “arches”in common (Fig. 1). These structures could also beimportant new characteristic features to distinguish

this subgroup. Considering the morphological fea-tures and the position in the phylogenetic tree,these two species would better to be included inthe family Medusettidae.

Wide Distribution of Phaeodarians

Specimens of phaeodarian were collected fromboth deep sea and surface (e.g. MS1–4, Table 1),emphasizing their wide vertical distribution as pre-viously reported (Nakamura and Suzuki 2015).Based on sampling location, C. diodon is likely acosmopolitan species. The two sequences of C.diodon within subclade F (Fig. 3) have been col-lected at distinct locations, Sogndalsfjord, westernNorway (NA [AB218765], Yuasa et al. 2006) and theEast China Sea (ECS4, Table 1). This species wasalso found in the North Sea, the Labrador Sea, theSouth Atlantic and the Mediterranean Sea (Borgert1901).

The present study provides a strong morpho-genetic framework for phaeodarian classification,and will be a valuable tool for accurate interpretationof ongoing environmental diversity surveys basedon DNA sequencing. This morphogenetic approachand the single-cell PCR method can further elu-cidate the intra- and inter-group relationships ofphaeodarians and other protists by combining otherbiological information such as distribution or physi-ology.

Methods

Sampling and identification: Plankton samples were col-lected in 2012 and 2013 from 14 stations located in 5 seasacross the northern hemisphere: the western and eastern NorthPacific Ocean, the Sea of Japan, the East China Sea andthe Mediterranean Sea (Supplementary Material Table 1, Sup-plementary Material Fig. S1). Phaeodarian individuals wereisolated manually from the samples under a stereomicroscopeor inverted microscope, as soon as possible after the samp-ling. Isolated specimens were put into wells of cell cultureplates and incubated for several hours to allow self-cleaning.The specimens were carefully identified according to their mor-phological characters under an inverted stereomicroscope andphotographed with a digital camera (DIGITAL SIGHT DS-Ri1,Nikon, Japan). For identification, all the documents concern-ing phaeodarian taxonomy were examined, and a taxonomicdatabase of synonyms of Phaeodaria was constructed. Thespecimens were then individually preserved in tubes filled withapproximately 2.0 mL of 99.9% ethanol. The tubes containingthe specimens were stored at 4 ◦C. Each specimen was labeledaccording to its sampling station (Table 1 and Fig. 1).

Single-cell PCR: After confirming that there were no otherorganisms attached to the cell surface, each specimen wasput into 50 �L of guanidine-containing extraction buffer (GITCbuffer, Decelle et al. 2012) and preserved in -80 ◦C overnight.Some large specimens (e.g. Aulosphaeridae) were dissectedby a sterilized scalpel, and only their “central capsules”

Molecular Phylogeny of Phaeodaria 371

Table 2. Sequences of the primers for 18S rDNA used in this study.

Primer name Position* Sequence 5′-3′ Direction Specificity Reference

Medlin A 5′-end AAC CTG GTTGAT CCT GCCAGT

Forward Universal Medlin et al.(1988)

PhaeoF3a 288–319 TCA TTC AAATTT CTG CCCTAT CAG CTWGAY GG

Forward Phaeodaria This study

NS2 546–566 GGC TGC TGGCAC CAG ACTTGC

Reverse Universal White et al.(1990)

Phaeo11F 559–583 CAG CAG CCGCKG TAA TTCCAG CTC C

Forward Phaeodaria This study

PhaeoF4-2 872–890 AKG GAT AGTTGG GGG TGC T

Forward Phaeodaria This study

Phaeo880F 872–893 AKG GAT RGTTGG GGG TGCTAG T

Forward Phaeodaria This study

Phaeo R 885–904 GGC CAY TRAATA CTA GCA CC

Reverse Phaeodaria Nakamura et al.(2013)

SR6 1761–1777 TGT TAC GACTTT TAC TT

Reverse Universal designed inVilgalys lab,Duke University.

Medlin B 3′-end CCT TCT GCAGGT TCA CCT AC

Reverse Universal Medlin et al.(1988)

*Position within 18S rDNA of Protocystis xiphodon

(protoplasmic body with nucleus) were transferred into the GITCbuffer. The specimens were heated at 70 ◦C for 20 min to breakthe wall of their central capsules. The extracted DNA was puri-fied with a chemagic DNA Plant kit (PerkinElmer chemagen,Germany). Single-cell PCR was conducted to amplify 18S rDNAwith a total reaction volume of 25 �L. A total of 9 primers wereused, including 4 phaeodarian-specific primers designed inthis study: Phaeo F3a, Phaeo11F, PhaeoF4-2 and Phaeo880F(Table 2). PCR reactions were performed using the followingprotocol: initial denaturation at 95 ◦C for 10 min, 40 cycles at95 ◦C for 20 sec, 58 ◦C for 15 sec and 72 ◦C for 120 sec with afinal extension at 72 ◦C for 7 min. The amplified PCR productswere purified with AMPure XP Kit (Beckman Coulter, USA).Sequencing reactions were conducted using an ABI PRISM3130xl Genetic Analyzer (ABI, U.S.A.).

Phylogenetic analysis: The sequences obtained wereassembled using ChromasPro (Technelysium Pty Ltd,Australia), and the alignments were checked manually.The sequences of 6 phaeodarian species, 57 cerco-zoans were also retrieved from the Basic Local AlignmentSearch Tool (BLAST) at NCBI (http://blast.ncbi.nlm.nih.gov/Blast.cgi)(Supplementary Material Table 2), and all thesequences were then aligned using MEGA 5 (Tamura et al.2011). Phylogenetic trees were inferred by maximum-likelihood(ML), neighbor-joining (NJ) and Bayesian analysis methods.For the ML and NJ trees, the General Time Reversible(GTR) model plus Gamma with the shape parameter foramong-site rate variation (G) was selected based on the

lowest Bayesian Information Criterion (BIC) score, and thesubstitution nucleotide matrix parameters were calculated. TheBootstrap values (Felsenstein 1985) were estimated basedon 1000 pseudo-replicates. GTR model with invariant sitesand the gamma distributed model (IG) were selected for theBayesian analysis, and the Bayesian posterior probabilities(BPP) were calculated with Mr Bayes version 3.2.2. (Ronquistand Huelsenbeck 2003). The Markov Chain Monte Carlochains ran for 3.0 x 107 generations, sampling every 1000generations. The first 1.0 x 106 generations were discarded asburn-in, checking by a program Tracer version 1.6 (Rambautand Drummond 2009). The remaining data reaching the steadystate were used for building the consensus tree, and the treewas visualized by Fig Tree version 1.4.0. (Rambaut 2012).

Acknowledgements

We thank the captains and crews of the T/SOshoro-maru (Hokkaido University), T/S Ushio-maru (Hokkaido University) and T/S Toyoshio-maru(Hiroshima University) for their assistance in samp-ling. We also thank the members of the oceanicplankton group of Station Biologique de Roscofffor their kind technical advice. Appreciation is also

372 Y. Nakamura et al.

due to the members of Plankton laboratory ofHokkaido University for their help in analysis. Thecomments of two anonymous reviewers and theeditor also greatly improved the manuscript. Thisstudy was financially supported by Grant-in-Aidfor JSPS Fellows (no. 26-2889, Y. Nakamura),by the Cooperative Research Project with Cen-tre National de la Recherche Scientifique (CNRS),“Morpho-molecular Diversity Assessment of Eco-logically, Evolutionally, and Geologically RelevantMarine Plankton (Radiolaria)” by the Strategic Inter-national Research Cooperative Program hosted bythe Japan Science and Technology Agency (JST)(N. Suzuki and F. Not), and by “Integrated Researchon Biodiversity of Interspecies Relationships” of theNational Museum of Nature and Science (A. Tuji).

Appendix A. Supplementary data

Supplementary data associated with this arti-cle can be found, in the online version, athttp://dx.doi.org/10.1016/j.protis.2015.05.004.

References

Adl SM, Simpson AG, Lane CE, Lukes J, Bass D, BowserSS, Brown MW, Burki F, Dunthorn M, Hampl V, Heiss A,Hoppenrath M, Lara E, Le Gall L, Lynn DH, McManus H,Mitchell EA, Mozley-Stanridge SE, Parfrey LW, Pawlowski J,Rueckert S, Shadwick RS, Schoch CL, Smirnov A, SpiegelFW (2012) The revised classification of eukaryotes. J EukaryotMicrobiol 59:429–493

Borgert A (1901) Die nordischen Tripyleen-Arten. Nord Plank-ton 15:1–52

Cachon J, Cachon M (1985) 2. Class Polycystinea. In Lee JJ,Hutner SH, Bovee EC (eds) Illustrated Guide to the Protozoa.Allen Press, Lawrence, pp 283–295

Campbell AS (1954) Radiolaria. In Moore RC (ed) Treatise onInvertebrate Paleontology, Part D, Protista 3. Protozoa (ChieflyRadiolaria and Tintinnina). Geological Society of America andUniversity of Kansas Press, Kansas, D1–D163

Decelle J, Suzuki N, Mahé F, de Vargas C, Not F (2012)Molecular phylogeny and morphological evolution of the Acan-tharia (Radiolaria). Protist 163:435–450

Felsenstein J (1985) Confidence limits on phylogenies: anapproach using the bootstrap. Evolution 39:783–791

Gowing MM, Coale SL (1989) Fluxes of living radiolarians andtheir skeletons along a northeast Pacific transect from coastalupwelling to open ocean waters. Deep-Sea Res 36:550–565

Haeckel E (1887) Report on the Radiolaria collected by H. M.S.Challenger during the years 1873–1876. Rep Sci Res VoyageHMS Challenger 1873–1876. Zool 18:1–1803

Howe AT, Bass D, Scoble JM, Lewis R, Vickerman K,Arndt H, Cavalier-Smith T (2011) Novel cultured protists

identify deep-branching environmental DNA clades of Cer-cozoa: new genera Tremula, Micrometopion, Minimassisteria,Nudifila, Peregrinia. Protist 162:332–372

Kling SA, Boltovskoy D (1999) Radiolaria Phaeodaria. InBoltovskoy D (ed) South Atlantic Zooplankton. Backhuys Pub-lishers, Leiden, pp 231–264

Kunitomo Y, Sarashina I, Iijima M, Endo K, Sashida K(2006) Molecular phylogeny of acantharian and polycystineradiolarians based on ribosomal DNA sequences, and somecomparisons with data from the fossil record. Eur J Protistol43:143–153

Lampitt RS, Salter I, Johns D (2009) Radiolaria: Majorexporters of organic carbon to the deep ocean. Global Bio-geochem Cycles 23, GB1010

Nakamura Y, Suzuki N (2015) Chapter 10 Phaeodaria:Diverse marine cercozoans of world-wide distribution. In:Ohtsuka S, Suzaki T, Horiguchi T, Suzuki N, Not F (eds),Marine Protists Diversity and Dynamics, Springer-Verlag, Berlin(in press)

Nakamura Y, Imai I, Yamaguchi A, Tuji A, Suzuki N (2013)Aulographis japonica sp. nov. (Phaeodaria, Aulacanthida, Aula-canthidae), an abundant zooplankton in the deep sea of the Seaof Japan. Plankton Benthos Res 8:107–115

Nikolaev SI, Berney C, Fahrni JF, Bolivar I, Polet S, Myl-nikov AP, Aleshin VV, Petrov NB, Pawlowski J (2004) Thetwilight of Heliozoa and rise of Rhizaria, an emerging super-group of amoeboid eukaryotes. Proc Natl Acad Sci USA 101:8066–8071

Pawlowski J, Holzmann M, Tyszka J (2013) New supraordinalclassification of Foraminifera: Molecules meet morphology. MarMicropaleontol 100:1–10

Pawlowski J, Christen R, Lecroq B, Bachar D, ShahbazkiaHR, Amaral-Zettler L, Guillou L (2011) Eukaryotic richnessin the abyss: insights from pyrotag sequencing. PLoS ONE6:e18169

Polet S, Berney C, Fahrni J, Pawlowski J (2004) Small-subunit ribosomal RNA gene sequences of Phaeodareachallenge the monophyly of Haeckel’s Radiolaria. Protist155:53–63

Rambaut A (2012) FigTree version 1.4.0. http://tree.bio.ed.ac.uk/software/figtree/

Rambaut A, Drummond AJ (2009) Tracer version 1.6.http://beast.bio.ed.ac.uk/Tracer

Ronquist F, Huelsenbeck JP (2003) MrBayes version 3. 0:Bayesian phylogenetic inference under mixed models. Bioinfor-matics 19:1572–1574

Sierra R, Matz MV, Aglyamova G, Pillet L, Decelle J, Not F, deVargas C, Pawlowski J (2013) Deep relationships of Rhizariarevealed by phylogenomics: A farewell to Haeckel’s Radiolaria.Mol Phylogenet Evol 67:53–59

Steinberg DK, Cope JS, Wilson SE, Kobari T (2008) A com-parison of mesopelagic mesozooplankton community structurein the subtropical and subarctic North Pacific Ocean. Deep-SeaRes II 55:1615–1635

Takahashi K (1991) Radiolaria: Flux, Ecology, and Taxon-omy in the Pacific and Atlantic. Ocean Biocoenosis Series 3.

Molecular Phylogeny of Phaeodaria 373

Woods Hole Oceanographic Institution Press, Woods Hole,303 p

Takahashi K, Anderson OR (2000) Class Phaeodaria. In LeeJJ, Leedale GF, Brandbury P (eds) Illustrated Guide to the Pro-tozoa. 2nd ed Society of Protozoologists, Kansas, pp 981–994

Takahashi K, Honjo S (1983) Radiolarian skeletons: size,weight, sinking speed, and residence time in tropical pelagicoceans. Deep-Sea Res 30:543–568

Tamura K, Peterson D, Peterson N, Steker G, Nei M, KumarS (2011) MEGA5: Molecular Evolutionary Genetics Analysisusing Maximum Likelihood, Evolutionary Distance, and Maxi-mum Parsimony Methods. Mol Biol Evol 28:2731–2739

Yuasa T, Takahashi O, Dolven JK, Mayama S, Matsuoka A,Honda D, Bjørklund KR (2006) Phylogenetic position of thesmall solitary phaeodarians (Radiolaria) based on 18S rDNAsequences by single cell PCR analysis. Mar Micropaleontol59:104–114

Available online at www.sciencedirect.com

ScienceDirect